A computational approach to optimal damping controller design for a GCSC

This paper presents a computational approach for an offline measurement-based design of an optimal damping controller using adaptive critics for a new type of flexible ac transmission system device-the gate-controlled series capacitor (GCSC). Remote measurements are provided to the controller to damp out system modes. The optimal controller is developed based on the heuristic dynamic programming (HDP) approach. Three multilayer-perceptron neural networks are used in the design-the identifier/model network to identify the dynamics of the power system, the critic network to evaluate the performance of the damping controller, and the controller network to provide optimal damping. This measurement-based technique provides an alternative to the classical linear model-based optimal control design. The eigenvalue analysis of the closed-loop system is performed with time-domain responses using the Prony method. An analysis of the simulation results shows potential of the HDP-based optimal damping controller on a GCSC for enhancing the stability of the power system.